Collective Intelligence for 2D Push Manipulations with Mobile Robots
This addresses the problem of enabling collective intelligence and adaptability in multi-robot systems for manipulation tasks, representing an incremental advance over prior restricted methods.
The paper tackles cooperative 2D push manipulations with mobile robots by distilling a planner from a differentiable physics simulator into an attention-based neural network, achieving better performance than baselines and generalizing to unseen configurations while adapting to environmental changes.
While natural systems often present collective intelligence that allows them to self-organize and adapt to changes, the equivalent is missing in most artificial systems. We explore the possibility of such a system in the context of cooperative 2D push manipulations using mobile robots. Although conventional works demonstrate potential solutions for the problem in restricted settings, they have computational and learning difficulties. More importantly, these systems do not possess the ability to adapt when facing environmental changes. In this work, we show that by distilling a planner derived from a differentiable soft-body physics simulator into an attention-based neural network, our multi-robot push manipulation system achieves better performance than baselines. In addition, our system also generalizes to configurations not seen during training and is able to adapt toward task completions when external turbulence and environmental changes are applied. Supplementary videos can be found on our project website: https://sites.google.com/view/ciom/home